摘要
该文针对基于地球物理的勘探方法不能有效地定量预测剩余油,提出了一种基于SOM网络的方法来预测剩余油。本文阐述了SOM网络的工作过程、算法的训练过程以及SOM网络聚类的方法。通过对含油饱和度敏感的属性进行聚类分析,可以及时预测剩余油的分布,对提高石油产量具有非常重要的现实意义。
This paper is aimed at the exploration methods which are based on geophysics can not predict the remaining oil accurately and quantitatively, a method which is based on SOM network is approached to predict the remaining oil. The paper briefly introduces SOM network work process, algorithm training process and cluster method. Through the sensitive attribute of oil saturation for clustering analysis, can predict the distribution of remaining oil in time which has very important practical significance to improve the oil production.
出处
《电脑知识与技术》
2016年第1Z期161-162,共2页
Computer Knowledge and Technology
关键词
剩余油
SOM网络
聚类分析
Remaining Oil
SOM Network
Clustering Analysis